Superalloys 718, 625, 706 and Derivatives 2005 Edited by E.A.

Loria
TMS (The Minerals, Metals & Materials Society), 2005

THE EFFECT OF COLD ROLLING ON THE CREEP BEHAVIOR OF

INCONEL ALLOY 718
C.J. Boehlert1, D.S. Dickmann2, and N.C. Eisinger3
1

Michigan State University, East Lansing, MI

2
Ferro Corporation, Rochester, NY
3
Special Metals Corporation, Huntington, WV
Keywords: microstructure, creep, backstress, creep rupture
Abstract
The creep behavior of INCONEL alloy 718 (IN 718) was investigated to identify processingcreep property relationships. The alloy was cold rolled (CR) to 0, 10, 20, 30, 40, 60, and 80%
followed by annealing and aging. In addition, this alloy can be superplastically formed (IN
718SPF) to a significantly finer grain size and the corresponding microstructure and creep
behavior were evaluated. The creep behavior was evaluated in the applied stress range of 300758MPa and the temperature range of 638-670C. Constant-load tensile-creep experiments were
used to measure the values of the steady-state creep rate and the consecutive load reduction
method was used to determine the values of backstress (o). Creep-rupture time (Tr) and
elongation-to-failure (f) were also evaluated at 649C and 758MPa. The lowest o values
(300MPa<0<310MPa) were exhibited for the most severely CR microstructures (60%, 80%,
and IN 718SPF), while the baseline 0%CR microstructure exhibited a significantly greater 0
value (540MPa). The greatest 0 values, 645 and 630MPa, were exhibited by the 20% and
30%CR conditions, respectively. The o values were related to the overall creep resistance as
the 20%CR condition exhibited the lowest secondary creep rates for a given applied stress (a),
while the samples CR to more than 40% exhibited the greatest creep rates. The values for the
effective stress exponent suggested that the transition between the rate-controlling creep
mechanisms was dependent on effective stresses (e=a-0) and the transition occurred at
e=135MPa for a temperature of 638C. Increased CR tended to increase Tr and f, and the
30%CR condition exhibited the greatest creep rupture properties. Overall, the 20%CR and
30%CR microstructures exhibited the greatest creep strength, while the most severely CR
materials exhibited the poorest creep strength.
Introduction
In order to identify processing-creep property relationships, the current work evaluated
the creep behavior of IN 718 as a function of sheet processing using 10% cold-rolling increments
from 0-80% followed by annealing and aging, and the tensile-creep properties were measured.
The approach taken was to characterize the creep behavior within the low-stress creep regimes
where diffusional creep and grain boundary sliding may be dominating the strain rate response,
and in addition the creep backstress was measured as a function of the amount of CR
deformation.

INCONEL is a registered trademark of Special Metals Corporation.

311

The creep mechanisms that operate during elevated-temperature deformation of pure

metals and solid-solution strengthened alloys have been related to the value of the stress
exponent, na, and the apparent activation energy, Qa, in the Dorn steady-state creep rate equation
[1]:

.
ss= A Do exp (-Qa/RT) b/kT (a/)

na

(1)

where T is the creep temperature in degrees Kelvin, R is the gas constant, and A is often referred
to as the Dorn constant, b is the Burgers vector, Do is the pre-exponential factor, is the shear
modulus, and k is the Boltzmann constant. However, the na and Qa values measured for alloys
containing dispersed second-phase particles are generally considerably greater than those
observed in pure metals and solid solution strengthened alloys [2-14]. The na values for
precipitation-hardened alloys have ranged between 5-15 [2-7] while, for dispersion-hardened
alloys, including thoria dispersed Ni-20Cr(wt.%), na values have ranged between 9-75
[10,13,14]. The Qa values have ranged from one to three times those of the activation energy for
self diffusion [12,14,15]. These variations in the observed values of na and Qa have been
rationalized by introducing the concept of a backstress or threshold stress, o, which is an
internal stress opposing the dislocation motion. In multiphase alloys such as IN 718, which
contains an austenitic FCC phase matrix () and fine and strengthening precipitates, the
applied stress during steady-state creep deformation is opposed by a backstress resulting from
the presence of these strengthening particles and a defect structure within the material [2,10,1621]. Therefore, the creep deformation results from an effective stress (e=a-o). As a result,
the steady-state creep rate can be represented by:

ne

ss = A*(a - o)

(2)

where ne is the effective stress exponent. Although traditionally used to measure the backstress
during high-stress dislocation power-law creep, it has been shown that the consecutive stress
reduction method can also be applied to low-stress diffusional creep for IN 718 where low ne
values are observed [16,17].
Experimental
The IN 718 sheets used in this study were processed at Special Metals Corporation,
Huntington, WV. The heats were produced by vacuum induction melting followed by
electroslag remelting. The material was hot worked using conventional practices and the asprocessed condition included mill annealing at 1066C for all hot-rolling procedures which
preceded the final CR and 982C anneal. Subsequent thermomechanical processing treatments
included CR between 10-80%. The CR steps were performed on separate sheets each designated
with 10%, 20%, 30%, 40,%, 60%, and 80% deformation. The annealing treatment was
performed at 954, 1010, or 1050C. In addition, one set of CR samples was heat treated below
the recrystallization temperature at 871C. The aging treatment, used to precipitate out the
and strengthening phases, consisted of 718C/8h/furnace cool to 621C then hold at 621C for
a total aging time of 18h. In addition, a separate heat of IN 718SPF was produced in a similar
fashion, however the sheet cold working procedure, estimated to total between 55-80%
deformation, was altered to assure the production of an ultrafine grain size product [22-25].
Flat dogbone-shaped tensile-creep specimens, with a cross-section of approximately
1mmx12mm and a gage length of 25mm, were machined, using either a mill or an
electrodischarge machine, with the tensile axis parallel to the rolling direction. Constant-load
creep experiments were performed on an Applied Test Systems, Incorporated (ATS) lever-arm
312

creep apparatus, using a 20:1 load ratio, in air at temperatures ranging between 638-670C and
applied stresses ranging between 300-758MPa. The creep strain was monitored during the tests
using a linear variable differential transformer attached to the gage section. The specimen
temperature, monitored by three thermocouples located within the gage section during the creep
experiments, was maintained within +3C using a single-zone ATS furnace. The o values were
determined at 638C by the consecutive stress reduction method [2,5,9]. When the creep rate for
a given a remained constant for at least five hours, it was assumed the steady-state creep rate
had been achieved. Thereafter, the sample was subjected to a small stress reduction
(approximately 5% a). This resulted in an elastic contraction of the sample, followed by an
incubation period with a zero creep rate. After a period of time, creep began again at a lower
rate. Once steady state was reached, another stress reduction was performed. The time of the
incubation period following each stress reduction was recorded. The remaining stress vs. the
cumulative incubation time was plotted, and o was determined by taking the asymptotic value
of the remaining stress when the cumulative incubation time appeared to be infinite. The o and
. values proved to be repeatable as duplicate samples were tested at the same temperature and
ss
.
.
a and the measured ss values were within five percent of each other. In addition, the o and ss
values were not dependent on strain history for total creep strains less than 0.5% as several
temperature/applied stress conditions were performed, some in duplicate, before that of the
backstress condition and in each case similar o values were recorded. Creep rupture
experiments were performed in air at 758MPa and 649C. All the creep experiments were
initiated after soaking the samples at the desired testing temperature for a minimum of 0.5 hours
to equilibrate the thermal stresses.
Results
Microstructure
The chemical composition range of the IN 718 heats used is shown in Table I. The annealed
microstructures contained an equiaxed -phase austenitic matrix and after aging fine (coherent
spherical fcc (L12)) and (coherent ordered disc-shaped body-centered tetragonal (DO22))
precipitated throughout, see Figure 1. The average grain diameter for the 0-40% CR sheets
ranged between 16-20m as measured through the line-intercept method. Thus 0-40% CR did
not drastically change the equiaxed grain size. However, the 60%CR, 80%CR, and IN 718SPF
microstructures exhibited an average grain diameter of 6.6m. Annealing temperature had a
significant effect on grain size. Above 1010C, grain growth occurs [26] and the 1050C, onehour annealed samples exhibited a grain diameter of 45m. It is noteworthy that the roomtemperature strength and hardness values significantly increased with CR deformation for the
871C heat-treated samples, while hardness remained almost constant after 954C annealing.
This indicated that 871C is below the austenite (-phase) recrystallization temperature for 040%CR IN 718, which is consistent with previous findings [27]. The relatively constant
hardness values for the 954C annealed samples indicated that the annealing temperature was
above the recrystallization temperature.
Table I. Composition range for the IN 718 heats used in this study
Element
Weight percent
Element
Weight percent

Ni
Ti
53.68-53.48 1.01-1.06
C
0.03

Mo
2.99

Fe
Cu
17.99-18.3 0.02-0.17

313

Co
0.03-0.12

Cr
18.1-18.4

Al
P
0.46-0.48 0.009-0.012

Si
0.01-0.17

Mn
Nb
0.04-0.12 5.07- 5.11

S
0.001

fine ppts

(a)
(b)
Figure 1. (a) Low-magnification and (b) high-magnification SEM photomicrographs of the
cross-section of a 0% cold rolled then 954C annealed-then-aged microstructure illustrating the
austenitic -phase matrix, fine and precipitates, and -phase precipitates (white).
Creep Behavior
During the creep experiments the material exhibited normal creep behavior, and the strain-time
.
plots illustrated the three stages of creep: primary, secondary, and tertiary. The dependence of ss
on a is illustrated in Figure 2. The na values, calculated from the slopes of these curves and
listed in Table II, were similar to those observed previously for IN 718 by Han and Chaturvedi
[17], whose data were interpolated to 638C and included in Figure 2 and Table II. Their data
was for an as-processed condition, and it closely resembled the 0%CR data in the current work.
Note that the na values, which ranged between 9-40, were similar to those measured for other
particle-strengthened alloy systems [2-14], and are considerably larger than those generally
observed for pure metals. It is apparent that there were two clusters of data, where the highest
strain rates were exhibited by the samples CR to more than 40% and lower strain rates were
exhibited by the 0-40%CR samples.
Figure 3 illustrates the plot used to determine o while Figure 4 illustrates the
.
corresponding
. ss versus e plot. The values of ne, listed along with ss and o in Table II, were
determined from equation 2. The 20%CR and 30%CR samples exhibited the greatest o values,
645 and 630MPa, respectively, while the 60%CR, 80%CR, and IN 718SPF conditions exhibited
the lowest o values, which were less than half those .of the highest o values. Correspondingly
the ss values for a given a were the lowest for the 20%CR and 30%CR materials, see Table II.
For e less than 135MPa, the ne values were between one and two, while for e greater than
135MPa, ne was greater than 3.5, see Table II and Figure 4. The Qa value (304+10kJ/mol) was
.
determined for
the
20%CR
material
using
the
ln

ss vs 1/T plot shown in Figure 5.

.

314

Table II. The 638C Creep Data of IN 718 samples annealed at 954C. then aged
Cold Rolling Deformation,%

0%

10%

20%

30%

40%

60%

80%

Han and Chatervedi [17] interpolated

to 638C

IN 718SPF

a, MPa

o, MPa

o - a, MPa

ss

na

ne

574
574
594
595
609
611
632
649
674
591
596
610
634
666
694
666
676
683
648
668
680
578
593
596
598
610
613
613
627
378
399
417
436
455
471
490
501
537
564
598
612
375
397
417
457
489
523
620
673
696
720
334
374
405

540
540
540
540
540
540
540
540
540
570
570
570
570
570
570
645
645
645
630
630
630
550
550
550
550
550
550
550
550
310
310
310
310
310
310
310
310
310
310
310
310
300
300
300
300
300
300
524
524
524
524
305
305
305

34
34
54
55
69
71
92
109
134
21
26
40
64
96
124
21
31
38
28
38
50
28
43
46
48
60
63
63
77
68
89
107
126
145
161
180
191
227
254
288
302
75
97
117
157
189
223
96
149
172
196
29
69
100

2.1E-09
1.9E-09
3.5E-09
3.8E-09
4.4E-09
4.6E-09
6.0E-09
9.1E-09
1.1E-08
1.0E-09
1.1E-09
2.8E-09
5.1E-09
1.1E-08
1.9E-08
3.8E-09
7.3E-09
1.0E-08
1.5E-09
5.6E-09
8.4E-09
2.4E-09
2.9E-09
2.8E-09
3.3E-09
4.7E-09
5.2E-09
4.8E-09
6.6E-09
3.3E-09
4.2E-09
5.2E-09
7.2E-09
9.9E-09
1.2E-08
2.0E-08
3.3E-08
5.8E-08
7.9E-08
1.2E-07
1.3E-07
2.3E-09
3.0E-09
3.8E-09
6.0E-09
9.7E-09
2.4E-08
5.9E-09
1.2E-08
1.7E-08
2.3E-08
2.1E-09
3.3E-09
6.0E-09

10.8
10.8
10.8
10.8
10.8
10.8
10.8
10.8
10.8
18.3
18.3
18.3
18.3
18.3
18.3
39.8
39.8
39.8
36.0
36.0
36.0
13.7
13.7
13.7
13.7
13.7
13.7
13.7
13.7
4.6
4.6
4.6
10.4
10.4
10.4
10.4
10.4
10.4
10.4
10.4
10.4
4.5
4.5
4.5
10.3
10.3
10.3
9.0
9.0
9.0
9.0
5.3
5.3
5.3

1.2
1.2
1.2
1.2
1.2
1.2
1.2
1.2
1.2
1.7
1.7
1.7
1.7
1.7
1.7
1.7
1.7
1.7
1.7
1.7
1.7
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.0
1.0
1.0
3.6
3.6
3.6
3.6
3.6
3.6
3.6
3.6
3.6
1.1
1.1
1.1
3.9
3.9
3.9
1.9
1.9
1.9
1.9
1.1
1.1
1.1

315

-7

638C

-1

Steady-State Creep Rate (s )

10

0% CR, na=10.8
10% CR, na=18.3

10

20% CR, na=39.8

-8

30% CR, na=36.0

40% CR, na=13.7
60% CR Low Stress, na=4.6
60% CR High Stress, na=10.4

10

-9

80% CR Low Stress, na=4.5

80% CR High Stress, na=10.3
SPF, na=5.3
Han and Chatervedi [33] Interpolated

-10

10

100

1000

10

Stress (MPa)

Figure 2. ss versus e plot for the 954C annealed-then-aged IN 718 creep samples tested at
638C. Also included is data from Han and Chaturvedi [33] which has been interpolated to
638C.

Remaining Stress (MPa)

300

0% CR
10% CR
20% CR
30% CR
40% CR
60% CR
80% CR

350
400

638C

450
500
550
600
650
700
0

20

40

60

80

100

120

140

Cumulative Incubation Time (hrs)

Figure 3. Remaining stress versus cumulative incubation time for the cold rolled and 954C
annealed-then-aged IN 718 creep samples.

316

-7

638C

-1

Steady-State Creep Rate (s )

10

-8

10

0% CR, ne=1.2
10% CR, ne=1.7
20% CR, ne=1.7
30% CR, ne=1.7
40% CR, ne=1.5
60% CR Low Stress, ne=1.0
60% CR High Stress, ne=3.6
80% CR Low Stress, ne=1.1
80% CR High Stress, ne=3.9

-9

10

-10

10

10

100

1000

Effective Stress (MPa)

Figure 4. ss versus e plot for the cold rolled and 954C annealed-then-aged IN 718 samples.
The data was used to calculate the listed ne values.

ln steady-state strain rate

-17
-17.2

=683MPa

-17.4

Qa=304 kJ/mole

-17.6
-17.8
-18
-18.2
-18.4
10.5
.

638C
670C
10.6

650C
10.7

10.8

10.9

11

10000/T (1/K)

Figure 5. ln ss vs 1/T plot used to calculate Qa for a 20%CR 954C annealed-then-aged sample.
The creep rupture data, listed in Table III, indicated that increased CR tended to increase
Tr. For the 954C annealed-then-aged samples, the greatest Tr and f values were exhibited by
the 30% and 40% cold rolled samples, where both the Tr and f values were greater than twice
those for the baseline 0% and 10%CR conditions. Cold rolling below 30% did not offer as
significant of an increase in the creep rupture properties and both the 10% and 20%CR

317

conditions resulted in lower f values than that for the baseline 0%CR condition. Each of the
ruptured samples exhibited ductile dimpling throughout the fractured surface. The 871C
annealed-then-aged samples tended to exhibit increased Tr values with increased CR and the Tr
values were significantly greater than the 954C annealed-then-aged samples. This is considered
to be a result, in part, to the greater tensile strengths exhibited by the 871C heat-treated
materials.
Table III. Creep rupture properties (649C/758MPa) for heat-treated (*954C,**871C) samples
Tr**, hr
Cold Rolling Deformation, % Tr*, hr
f , * %
f, ** %
0
24.9
9.3
73.6
26.1
10
27.5
6.7
65.7
25.2
20
38.5
7.9
99.8
22.0
30
64.8
28.9
103.8
18.2
40
57.9
38.0
128.9
23.8
Discussion
Creep Behavior
Effective stress exponent According to Wilshire and coworkers [2,4,5] and Gibeling and Nix
[19], the consecutive stress reduction method can be used to measure o during dislocation
power-law creep where an incubation time after a small stress reduction is observed to occur.
However, for IN 718 Han and Chaturvedi [16,17] observed an incubation time in both the
power-law and the diffusional-creep regimes. According to Harris [28] and Burton [29], for
diffusional creep to continue, the stress concentration at the precipitate-matrix interface created
by the entrapment of diffusing vacancies can only be relieved by the formation of prismatic
dislocation loops. However, Ansel and Weertman [30] have suggested that diffusional creep in
two-phase alloys can only occur by the process of dislocation climb over the precipitate particles.
Therefore diffusional creep in two-phase alloys may not only involve vacancy diffusion in the
matrix and dislocation motion in the grain boundary region but also dislocation creation and
motion within grains. Similar to the process of dislocation slip during power-law creep,
dislocation climb during diffusion creep also requires dislocation networks or line segments to
grow to a sufficient length so that they can be activated. These dislocation segments which
become active are known as Bardeen-Herring sources [31]. Han and Chaturvedi [16] observed
dislocation networks within grains where it was suggested that the observed incubation time
could be due to the activation of these Bardeen-Herring sources. Therefore it was concluded that
the consecutive stress reduction method can be used to determine o during power-law creep as
well as diffusional-creep.
The experimental results of the current work indicate two separate regimes based on the
effective stress level. The ne values were measured to be between one and two for e less than
135MPa, and ne was 3.6-3.9 for e greater than 135MPa. Based on this result, the creep
mechanism may be dependent on e, and the low-stress regime (a<450MPa) may be dominated
by diffusional creep or grain boundary sliding. The mechanism of diffusional creep in particlestrengthened alloys is different from that of single-phase alloys or pure metals. Diffusional creep
in the latter is known to be the result of stress-induced diffusion or the migration of matter from
grain boundaries that are in tension to those that are in compression. In contrast, in particlestrengthened alloys, diffusion of vacancies will be inhibited by the precipitate particles with an
accumulation of vacancies at the precipitate-matrix interface and a resultant build-up of stress
concentration. This stress concentration at the interface can be relaxed by the punching of
prismatic dislocation loops in the matrix, which can annihilate themselves by the absorption of
318

vacancies [28,29]. Therefore, to accommodate even very limited diffusional strain, in particlestrengthened alloys, additional plastic deformation must occur around precipitates either in the
grain boundary region or within grains, giving an ne value slightly greater than 1. Furthermore,
diffusional creep can involve the process of dislocation climb over the particles [30]. Ansel and
Weertman [30] considered the rate controlling process to be the climb of dislocations over the
precipitate particles, which also depends on the diffusion rate. However, if cross-slip is also
important in the above process, the creep rate will be more sensitive to stress. Han and
Chaturvedi [17] observed dislocation segments and loops within grains even under the testing
conditions where they considered diffusional creep to occur. Their observations suggest that the
creep is more sensitive to the applied stress in particle-strengthened alloys than to that in pure
metals; i.e. the value of ne can be slightly greater than 1. This observation is in agreement with
the findings of Ansel and Weertman [30], and it may also help explain the ne values of the
current work.
Activation energy The measured Qa value (304kJ/mol) was greater than both the activation
energy for self diffusion (265-280kJ/mol) and the activation energy for the creep process
(276kJ/mol) of pure Ni [32] and the creep process of Ni-Cr in solid solution (295kJ/mol) [33].
However, the creep rate expression used to calculate these values considers neither the influence
of temperature on the value of G nor the concept of backstress. Using the backstress, lower
activation energies than those calculated using the applied stress have been calculated [17], and
such values lie within the range expected for lattice self diffusion. Thus a similar result would be
expected based on the Qa measured here, and self-diffusion is considered to be more likely than
grain boundary diffusion for creep of IN 718 in the temperature range of 638-670C
Backstress The significant drop in o with increased CR deformation beyond 40% was
correlated with a decreasing average grain diameter of approximately 20m for the baseline
0%CR material to approximately 6m for the 60%CR, 80%CR, and IN 718SPF microstructures.
The o values observed in the current work suggest that o may be dependent on grain size. At
this point, the reason for the dramatic increase in backstress for the 20%CR and 30%CR
conditions is not apparent and observations of the deformed samples are necessary.
Creep rupture Increased CR tended to increase Tr and f. The IN 718SPF material was not
evaluated in creep in this study, but based on previous creep rupture data [22] its Tr value is
expected to be similar to that exhibited by the 20%CR condition. Thus there appears to be a
limit to the amount of CR deformation that will result in increased creep rupture life and f. This
limit appears to be near 30%CR as this condition exhibited the maximum Tr value and a decrease
in Tr was observed at 40%CR for the 954C annealed-then-aged samples. It is noted that a
significant decrease in the grain size occurred with increased CR from the 40%CR condition to
the IN 718SPF condition, and this may be a significant factor in the creep rupture discrepancy.
Comparing the creep strain rate and rupture properties, it appears that the 30%CR condition
results in the most attractive overall creep behavior.
Summary
IN 718 was processed through sequential increments of CR between 0-80% followed by
annealing and aging to evaluate processing-creep property relationships. The steady-state creep
rate, backstress, and creep-rupture properties were measured for 954C annealed-then-aged
samples. The greatest backstress and lowest creep rate values were exhibited by the 20% and
30%CR microstructures. The lowest backstress and greatest creep rates were exhibited by the
most severely CR microstructures which exhibited the finest grain size. The effective stress

319

exponent values, which incorporated the backstress, suggested that the creep deformation
mechanism is dependent on effective stresses (e) where the transition point occurs at
e~135MPa. Trends in the creep rupture data indicated that both Tr and f increases with
increased CR. Overall, the 30%CR condition exhibited the most attractive creep properties.
Acknowledgments
This work was supported by the NSF Division of Materials Research (DMR-0134789). The
authors acknowledge Gaylord Smith (SMC) and James Crum (SMC) for helpful guidance.
References
1. A.K. Mukherjee, J.E. Bird, and J.E. Dorn: Trans. Am. Soc. Metals, 1969, vol. 62, pp. 155-79.
2. K.R. Williams and B. Wilshire: Met. Sci. J., 1973, vol. 7, pp. 176-9.
3. D. Sidney and B. Wilshire: Met. Sci. J., 1969, vol. 3, pp. 56.
4. J.D. Parker and B. Wilshire: Met. Sci. J., 1975, vol. 9, pp. 248.
5. P.W. Davies, G. Nelmes, K.R. Williams, and B. Wilshire: Met. Sci. J., 1973, Vol. 7, pp 87-92.
6. W.J. Evans and G.F. Harrison: Met. Sci. J., 1976, vol. 10, pp. 307.
7. R. Lagneborg and B. Bergman: Met. Sci. J., 1976, vol. 10, pp. 20.
8. D.D. Sherby and P.M. Burke: Prog. Mater. Sci., 1967, vol. 13, pp. 325.
9. J.D. Parker and B. Wilshire: Metal Science, October 1978, pp. 453-8.
10. R. Lund and W.D. Nix: Acta Metall., 1976, vol. 24, pp. 469.
11. C.N. Ahlquist and W.D. Nix: Acta Metall 1971, vol. 19, pp. 373-85.
12. C.L. Meyers, J.C. Shyne, and O.D. Sherby: Aust. Inst. Met., 1963, vol. 8, pp. 171.
13. B.A. Wilcox and A.H. Clauer: Trans. Metall. Soc. AIME, 1966, vol. 236, pp. 570
14. A.H. Clauer and B.A. Wilcox: Met. Sci. J., 1967, vol. 1, pp. 86.
15. B.A. Wilcox and A.H. Clauer: Met. Sci. J., 1969, vol. 3, pp. 26.
16. Y. Han and M.C. Chaturvedi: Mater. Sci. Eng., 1987, vol. 85, pp. 5965.
17. Y. Han and M.C. Chaturvedi: Mater. Sci. Eng., 1987, vol. 89, pp. 2533.
18. W. Chen and M.C. Chaturvedi: Materials Science and Engineering, 1994, vol. A183, p 81-9.
19. J.C. Gibeling and W.D. Nix: Materials Science and Engineering, 1980, vol. 45, p.123.
20. S. Purushothman and J.K. Tien: Acta Materialia, 1978, vol. 26, pp. 519.
21. J.H. Hausselt and W.D. Nix: Acta Materialia, 1977, vol. 25, pp. 595.
22. G.D. Smith and D.H. Yates: Proc. Advancements in Synthesis and Processes, 1992, Society
for the Advancement of Material and Process Engineering, Covina, CA, pp. M207-M218.
23. G.D. Smith and H.L. Flower: Proc. Superalloys 718, 625, 706 and Various Derivatives,
1994, The Minerals, Metals and Materials Society, Warrendale, PA, pp. 355-364.
24. B.A. Baker: INCO Alloys International Technical Investigation Report No. BAB1323093,
Huntington, WV. September 1993.
25. Y. Huang and P.L. Blackwell: Materials Science and Technology, 2003, Vol. 19, pp. 461-6.
26. INCONEL alloy 718 Bulletin, 4th Edition, 1985, published by The International Nickel
Company, Inc., now Special Metals Corporation, p.1-25
27. W.C. Liu, Z.L. Chen, and M. Yao: Metall. and Materials Transactions, 1999, 30A, p 31-40.
28. J.E. Harris: Met. Sci. J., 1973, vol. 7, p.1.
29. B. Burton: Materials Science and Engineering, 1973, vol. 11, pp. 337.
30. G.S. Ansel and J. Weertman: Trans. Metall. Soc. AIME, 1959, vol. 215, p.838.
31. D. Hull: Introduction to Dislocations, Pergamon, Oxford, 1975, pp.188-190
32. J.P. Dennison, R.J. Llewellyn, and B. Wilshire: J. Inst. Met., 1967, vol. 95, p.115.
33. D. Sidey and B. Wilshire, Met. Sci. J., 1969, vol. 3, p.56.

320